Last updated: June 2026
Azure Blob reserved capacity is a one or three year commitment to a fixed quantity of stored data, bought in 100 TB or 1 PB blocks, in exchange for a lower per-gigabyte storage rate, up to roughly a third below pay-as-you-go on the covered charge.
- Buy it for the stable floor of your stored data, never your peak. The commitment is use-it-or-lose-it.
- It covers the storage charge only. Transactions, bandwidth, and early-deletion fees stay at pay-as-you-go.
- It does not lock a tier. Hot, Cool, Cold, and Archive usage in scope all draw on the same reservation.
- The deepest discounts are on 1 PB blocks and 3-year terms.
Azure Blob reserved capacity is a billing commitment: you agree to keep a set amount of data, measured in 100 TB or 1 PB units, in Blob Storage for one or three years, and Azure charges you a discounted per-gigabyte rate for that committed amount. It is the storage equivalent of a reserved instance, applied to bytes at rest rather than compute hours. This article is part of our Azure cluster; start with the complete guide to Azure cost optimization, the pillar this piece links up to. Buying storage commitments is a Lock step in our See, Cut, Lock, Run method, and the order matters: clean up and tier the data first, then reserve the clean baseline.
What is Azure Blob reserved capacity?
Azure Blob reserved capacity is a discount you unlock by committing to a fixed volume of block blob storage for a one or three year term. Instead of paying the standard per-gigabyte rate every month, you pre-commit to a quantity and pay a reduced rate on it, either upfront or in equal monthly payments at the same total cost. Microsoft sells it in two block sizes, 100 TB and 1 PB, and you select the storage tier, the redundancy, and the region the reservation applies to. The covered savings run up to roughly a third versus pay-as-you-go on the storage charge, per Microsoft's Blob reserved capacity documentation, with larger discounts on the 1 PB block and the 3-year term.
What does reserved capacity actually cover, and what does it not?
Reserved capacity discounts the data-at-rest storage charge and nothing else. That is the single most misunderstood point and the one that trips up the savings math. The reservation applies to the per-gigabyte cost of the bytes you store. It does not touch the other lines on a storage bill.
| Charge | Covered by reservation? | Notes |
|---|---|---|
| Stored data per GB | Yes | The committed quantity is billed at the discounted rate |
| Transactions and operations | No | Read, write, list, and other operations stay at pay-as-you-go |
| Bandwidth and egress | No | Data leaving Azure is billed separately |
| Early deletion of Cool/Cold/Archive | No | Minimum-retention penalties still apply |
| Index and metadata features | No | Optional features bill at their own rates |
This is why a storage cleanup belongs before the reservation. If you reserve a petabyte that includes stale snapshots, orphaned blobs, and old log data nobody reads, you lock in a discount on waste. Clear the waste first, see what stable volume remains, then commit to that. The same sequence applies across every commitment, which is why we treat Azure reserved capacity for SQL and Cosmos DB with the same discipline.
How are the 100 TB and 1 PB units priced?
Reserved capacity comes in two block sizes, and the larger block carries the deeper discount. You buy whole units, so the practical floor is 100 TB committed. If your stable footprint is 340 TB, you might reserve three 100 TB blocks and let pay-as-you-go cover the remaining variable 40-plus TB. If it is several petabytes, the 1 PB blocks give both a cleaner fit and a better rate. The reservation is scoped to a tier, a redundancy option such as LRS, ZRS, or GRS, and a region, so a single account that spans regions or redundancy types may need more than one reservation to cover its footprint.
Reserve the floor, not the peak. Look at twelve months of stored-data history, find the lowest stable level the account never drops below, and commit to that. Pay-as-you-go absorbs everything above the floor at no penalty, while every gigabyte you reserve below the floor is discounted with zero waste risk.
Does reserved capacity lock me into a storage tier?
No. The reservation commits you to a quantity of data, not to a specific access tier, so lifecycle management still works normally. You can keep moving data from Hot to Cool to Cold to Archive as it ages, and the reservation continues to apply to the matching covered usage. This matters because tiering and reserving are complementary, not competing, levers. Tiering reduces the rate per gigabyte by access frequency; reserving reduces it by commitment. Used together on a clean dataset, they compound. To get the tiering layer right first, pair this with the storage choices in Azure Files vs Blob: choosing the cheaper storage service.
Sitting on petabytes of Blob you have never reserved?
Our Azure cost audit cleans and tiers your storage first, finds the stable floor worth committing, and sizes the reservation so you discount the baseline without stranding the commitment. On the performance model, you pay only from realized savings. No savings, no fee.
Book an Azure cost audit →When should you buy reserved capacity, and when should you wait?
Buy when your stored-data volume is large and stable, because that is exactly the profile the discount rewards. A few hundred terabytes or more, held for the foreseeable future, with a clear floor, is the textbook case. Wait when the data is small, when the volume is volatile or trending down, or when you have not yet cleaned and tiered it. A reservation on data you are about to delete is a loss, and a reservation sized to a peak you rarely hit wastes part of the commitment every month. The decision is the same shape as any other commitment question, which is why we model it inside the broader Azure cost optimization program rather than as a one-off purchase.
Pricing, block sizes, tier coverage, and discount percentages reflect Azure Blob reserved capacity as of June 2026. Verify the current per-unit rates and exact discount for your tier, redundancy, and region in Azure Blob Storage pricing before committing, because storage pricing changes.
The Azure Cost Optimization Field Guide includes the storage cleanup checklist and the commitment sizing worksheet we apply on engagements. It is the downloadable companion to this article.
Frequently asked questions
Does Azure Blob reserved capacity lock me into a storage tier?
No. Reserved capacity is a commitment to a quantity of stored data, not to a tier. The discount applies to Hot, Cool, Cold, and Archive usage that matches the reservation scope, so moving data between access tiers through lifecycle policies does not break the reservation.
What does Azure Blob reserved capacity not cover?
Reserved capacity covers the per-gigabyte storage charge only. Transaction and operation costs, early-deletion fees, bandwidth, and data-transfer charges are billed separately at pay-as-you-go rates and are not discounted by the reservation.
How much can Azure Blob reserved capacity save?
Microsoft documents savings of up to roughly a third versus pay-as-you-go on the covered storage charge, with the largest discounts on 1 PB blocks and 3-year terms. Verify the exact percentage for your tier, region, and redundancy in current Azure pricing before committing.
What happens if I store less than my reserved amount?
You still pay for the full reserved quantity. The reservation is a use-it-or-lose-it commitment, so size it to the floor of your stable stored data, not your peak, and let pay-as-you-go absorb the variable layer on top.
The short version
Azure Blob reserved capacity discounts the storage charge on a committed 100 TB or 1 PB block for one or three years, saving up to roughly a third on the covered usage, with the best rates on the largest blocks and longest terms. Clean and tier the data first, reserve the stable floor, and remember the reservation covers stored bytes only. When you want the cleanup, the tiering, and the commitment sized in one pass, that is exactly what our Azure cost optimization service delivers.
Cloud pricing and service behavior change frequently. Verify the specifics in this guide against the providers’ own current documentation and the FinOps Foundation: Azure pricing ↗, Azure documentation ↗ and FinOps Foundation Framework ↗. This article also reflects Cloud Cost Room’s hands-on, vendor-neutral engagement experience.